Home |  Electronics |  Toolbox |  Gini and Karl's world |  E-cards |  Photos |  Online-Shop



Content:
By Guido Socher
<guido_at_tuxgraphics_org>

 

wind vane, first version

[Illustration]

Abstract:

Traditional wind measurement equipment for weather stations consists of a wind vane (wind direction) and a cup anemometer (wind speed). But it's kind of boring to just build what others have done already. There are also some mechanical problems:

  • A good wind vane and cup anemometer would require water proof ball bearings. They are not easy to get.

  • It is not so straight forward to build a sensor that can measure an angle and be turned around infinetly. Something like that would be required for the wind vane. Proffesional wind vanes seem to use a potentiometer which can be turnded entlessly. ...but where to buy such a special potentiometer?

This it was very interessting to build this wind direction meter. I observed a few things that I did not expect at the beginning. It failed however because the switches which I used failed after just 3 month. The actual idea was not that bad -- it worked.
_________________ _________________ _________________

 

The idea

It is probaly not important to measure exactly from which angle the wind is comming. Just an indcation of noth, west, east, south and one value inbetween would be good enough. This can be done with just 4 switches. An inbetween value, e.g north-west is detect when both the switch for north and west are pressed. My idea was to build something like this:

A bottle as wind "wind vane". The picture shows the situation without wind. With wind from the right the bottle would tilt to the left.

The added benefit of this equipment would be that it will not only indicate the direction from which the wind is comming but also give no reading when there is no wind. The switches have to "soft enough" to trigger already at very little wind but that should not be a problem.  

Building it...

This is how I made that "bottle wind vane".


4 dip switches and a long screw in the middle. There are some rubber washers under the screw. The screw can therefore tilt a bit.


How the screw tilts thanks to the rubber washers.


A second board which will press the switches. It will also protect the switches from rain.


Side-view.


Mount it on a solid wooden board.


The bottle holder and a plastic foil for better protection of the electrical parts against rain.


It's ready.


Mount it outside on a high point (4m above ground).

 

The outcome

This project was successful because the idea as such worked. Just the switches failed. Why was that?

The interssting thing about this bottle wind vane was that it did not quite behave as I expected. I though that once the wind is strong enough to tilt the board and click a switch it will stay that way for a longer period of time. This is however not the case. The wind is never constant or steady. Even strong wind comes in short waves.

With very little wind the switches would be pressed occasionally and with stong wind the switches would be pressed very frequently. On a windy day the switch oposite to the wind direction would be pressed a 100000 times. Winds come here mostly from the west. The switch indicating west-wind failed therefore after 3 windy winter month.

I did not expect that the wind comes near ground (4m above ground) always in such waves. It is kind of logical to expect this if you remember how trees swing back and forth in the wind but I had never really tought about this. Tranditional wind speed measurements never show that. A wind mill or a cup anemometer will always measure just an average wind speed.

At the same time this is a very good behaviour because I could use that equipment to measure also wind speed. Not exact wind speeds but something along the terms of "low wind" or "strong wind".  

Conclusion

It failed because switches are not durable enough but the idea as such was good. I will re-desing it and find a replacement for the switches.



© Guido Socher, tuxgraphics.org

2007-04-02, generated by tuxgrparser version 2.55